The limits of min-max optimization algorithms: convergence to spurious non-critical sets.
Ya-Ping HsiehPanayotis MertikopoulosVolkan CevherPublished in: CoRR (2020)
Keyphrases
- min max
- optimization problems
- max min
- discrete optimization
- computational complexity
- stochastic search
- global convergence
- combinatorial optimization
- global optimization
- learning algorithm
- computational cost
- neural network
- stopping criteria
- stochastic approximation
- optimization methods
- convergence rate
- data structure
- times faster
- test problems
- optimization procedure
- orders of magnitude
- multiobjective optimization
- faster convergence
- machine learning algorithms
- optimization algorithm
- worst case
- significant improvement
- convergence analysis
- quasi newton
- data sets
- approximately optimal